The Rise of Conversational BI and NLP's Impact : a Systematic Literature Review

Abstract

This systematic literature review explores the impact of Natural Language Processing (NLP) in developing Business Intelligence (BI) systems focusing on the rise of Conversational Business Intelligence (CBI). It seeks to determine how NLP can improve user accessibility, decision making, and options available in navigating integration concerns in BI frameworks. Using the PRISMA 2020 guidelines, the review examined 18 peer-reviewed studies presented in the period between 2019 and 2024 through the Google Scholar and the Saudi Digital Library. Inclusion criteria based on pre-set criteria of NLP’s utilization in BI were applied to studies, and for their quality – methodological rigor and relevance, were considered. Findings had to be thematically grouped to handle issues of user accessibility, decision consequences and technical issues. NLP obviously increases BI accessibility with conversational interfaces that empower non-technical users, up to 30% more adoption rates in self-service systems. It enhances decision making using advanced analytics; sentiment analysis (85% accuracy) and predictive modeling (>95% accuracy) enable real time insights. However, scalability limitation, computational requirement and ethical issues such as bias and privacy call for strong solutions for CBI’s effective deployment. NLP integration of BI systems creates transformative value in terms of organizational data application, but facing technical and ethical challenges, adoption is not an easy task. In future research, building of scalable architectures, domain-specific NLP applications and use of ethical frameworks should be considered for CBI systems to be accessible, efficient and trustworthy. These have an implication that calls for interdisciplinary activities in ensuring that technological innovation is matched with practical utility.